3D facial images reconstruction from single facial image

摘要

3D facial images are important in many computer vision and graphic applications, such as facial recognition and animation. However, it is difficult to establish 3D images since it takes much cost and time. Thus, in my research, we try to reconstructed 3D images from only one facial image using machine learning approach. The results show that the proposed method can effectively generate the 3D images using only one input 2D image of 2816 by 2112in size, and the average error is 8 pixel.